± HEALTH Colorectal Cancer (CRC) is a common cause of cancer death and prognostic factors are used to determine manage…+ HEALTH This multicenter study developed and validated an interpretable machine learning model integrating granular n…+ HEALTH Multiple stakeholders need to locate results of registered clinical trials but frequently struggle to find th…+ HEALTH Accurate preoperative glioma grading and molecular subtyping are important for treatment. The vascular microe…+ HEALTH Gastrointestinal endoscopy generates extensive high-resolution video data, posing significant challenges for…+ HEALTH Non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) and invasive encapsulat…+ POLICY Artificial intelligence (AI) has the potential to strengthen real-world evidence (RWE) for regulatory decisio…± HEALTH Echocardiography remains the cornerstone of cardiovascular imaging. However, traditional workflows including…
TruaceTracing the truth around AIFriday, August 28, 2026
Subscribe
Health·G Space·Evidence-backed gain·Published 2026-08-28

TrialScout links published results to trial registrations using a large language model

Abstract: Multiple stakeholders need to locate results of registered clinical trials but frequently struggle to find them. Summary results of clinical trials are often not published in trial registries, and publications containing trial results are often not explicitly linked to their respective trial registrations. Finding these results is important to researchers, systematic reviewers, research funders, regulators, clinical practitioners, and patients. We developed TrialScout, a computer program that uses a large langua…

TRV-2026-0918Peer-reviewedPermanent record — cite & verify
TrialScout links published results to trial registrations using a large language model

PUBLICATION AND DISCLOSURE ISSUES IN ANTIDEPRESSANT PEDIATRIC CLINICAL TRIALS by Committee on Energy and Commerce. Public domain

The quick read

On 2026-08-26, a peer-reviewed study described TrialScout, a program that uses a large language model to link ClinicalTrials.gov registrations to PubMed result publications. Tested against prior human-coded datasets, it achieved 92.5% sensitivity and 81.2% specificity, and when applied to 9,600 sampled completed or terminated trials it identified publications for 6,110 trials.

The work matters because locating trial results is critical for evidence synthesis, regulatory oversight, and clinical decision-making, and automation could speed monitoring of reporting practices. Uncertainty remains about true accuracy because no gold standard exists, and the reported 61.5% human-error rate in disagreements suggests evaluation itself is imperfect.

Main points
  • TrialScout was developed to match clinical trials registered on ClinicalTrials.gov with corresponding result publications indexed in PubMed.
  • Performance was evaluated against human-coded matches from previous studies of results reporting rates.
  • Manual review of 200 disagreements found a majority (123/200, 61.5%, 95% CI, 54.4-68.3%) were due to human errors.
  • Cross-sectional analysis applied TrialScout to a random sample of 9,600 completed or terminated trials.
Gain

TrialScout using a large language model matched registered trials to result publications with 92.5% sensitivity and located publications for 63.6% of a 9,600-trial sample, accelerating locating trial results.

The rundown

Researchers built TrialScout to address the gap where summary results are often not published in registries and publications are often not explicitly linked to registrations, affecting researchers, systematic reviewers, funders, regulators, practitioners, and patients.

Evaluation compared TrialScout to human-coded matches, then applied it to 9,600 completed or terminated trials from ClinicalTrials.gov, with manual review of 200 discordant cases to adjudicate errors.

Sources

Reader signal

How should this claim be treated?

The debate